EARTHWORKS Study: Gas Patch Routlette - How Shale Gas Development Risks Public Health in Pennsylvania

Gas Patch
Roulette
HOW SHALE GAS
DEVELOPMENT RISKS PUBLIC
HEALTH IN PENNSYLVANIA
OCTOBER 2012

TM
EARTHWORKS

TM
EARTHWORKS

TM
EARTHWORKS
OIL & GAS ACCOUNTABILITY PROJECT

A WORD OF THANKS

A goal of this research project has been to give voice to the many people in
Pennsylvania (and beyond) who directly bear the costs of the nation’s dependence
on fossil fuels.
This report reflects the tremendous concern, caring, and openness of the project
participants. Thank you for giving your time, sharing what are often difficult and
personal experiences, and trusting us to write about them.
We are also grateful to the many people from local communities and partner
organizations who provided contacts and guidance. Special appreciation
goes to the Southwest Pennsylvania Environmental Health Project, which
provided advice early on in the research process and reviewed this report.
Finally, many thanks to the Colcom Foundation for its generous
support of this project and commitment to protecting the
environment and public health.

Photo by: Robert Donnan

Gas Patch Roulette
HOW SHALE GAS DEVELOPMENT
RISKS PUBLIC HEALTH IN
PENNSYLVANIA
OCTOBER 2012

By
Nadia Steinzor, Earthworks’ Oil & Gas Accountability Project
Wilma Subra, Subra Company
and Lisa Sumi, environmental research and science consultant

For more information on this study go to: http://health.earthworksaction.org

TM
EARTHWORKS

Cover photo: Mark Schmerling; inset photos L-R: Frank Finan, Mark Schmerling
Back cover photo: Nadia Steinzor; inset photos L-R: Roberto M. Esquivel/Herald-Standard, Nadia Steinzor

TM
EARTHWORKS

Earthworks’ Oil & Gas Accountability Project
Offices in New York State, Maryland, New Mexico, Colorado, Texas and California
EARTHWORKS • 1612 K St., NW, Suite 808 Washington, D.C., USA 20006
www.earthworksaction.org • www.ogap.org • 202.887.1872
OGAP
TM
EARTHWORKS

Table of Contents

INTRODUCTION ....................................................................................................................... 5

1. BACKGROUND: THE MARCELLUS SHALE BOOM ............................................................. 7

2. THE STUDY ........................................................................................................................... 8

2.1 OVERVIEW .................................................................................................................................................. 8
2.2 METHODOLOGY....................................................................................................................................... 8
2.3 FINDINGS .................................................................................................................................................. 10
PARTICIPANT OVERVIEW .............................................................................................................. 10
HEALTH SYMPTOMS ...................................................................................................................... 11
ODOR EVENTS .................................................................................................................................. 20
2.4 ENVIRONMENTAL TESTING ................................................................................................................ 21
AIR ......................................................................................................................................................... 21
WATER ................................................................................................................................................. 25
SYMPTOM AND TESTING ASSOCIATIONS .............................................................................. 28
2.5 CONCLUSIONS ........................................................................................................................................ 31

3. MISSING PIECES ................................................................................................................ 32

3.1 SCIENCE AND TESTING ........................................................................................................................ 32
3.2 POLICY AND REGULATION ................................................................................................................. 34

4. RECOMMENDATIONS ....................................................................................................... 35

5. FINAL WORDS .................................................................................................................... 37

6. REAL PEOPLE, REAL LIVES ................................................................................................ 39

GAS PATCH ROULETTE: HOW SHALE GAS DEVELOPMENT RISKS PUBLIC HEALTH IN PENNSYLVANIA 4
Earthworks’ Oil & Gas Accountability Project • www.earthworksaction.org

Introduction
Where oil and gas development goes, health problems often follow.

For many people across the United States, this statement has rung painfully true for a long time. As
the drilling boom picks up speed and reaches more places, it is now resonating in new
communities. From a growing number of stories told by individuals nationwide1 to conferences
held by academics and public agencies,2 the “dots” between health symptoms and gas facilities are
very slowly but surely being connected.

The health survey and environmental testing project described in the following pages is part of
this critical process. Between August 2011 and July 2012, Earthworks’ Oil & Gas Accountability
Project (OGAP) investigated the extent, types, and possible causes of health symptoms
experienced by people living in the gas patches of Pennsylvania.

Founded in 1988, Earthworks is dedicated to protecting
communities and the environment from the impacts of
irresponsible mineral and energy development while seeking Though knowledge of impacts
sustainable solutions. We reform government policies, evolves slowly, gas and oil
improve corporate practices, and work with landowners, extraction and production continue
organizations, agencies, and elected officials to adopt to accelerate rapidly—allowing
policies to protect public health and the environment and industry to put still-emerging
hold industry accountable for its practices. technologies to use without first
establishing their safety.
The findings of this study stand in strong contrast to
statements—often made by industry representatives and
policymakers seeking to expand drilling—dismissing claims
of health impacts as “personal anecdotes” and isolated
incidents. Directly impacted people are frequently told that
what they experience is a random occurrence and that some
other source—traffic, lifestyle choices, family disease history,
household products—is to blame.

We know that the gas and oil industry uses toxic substances
that harm human health. For example, of about 300
compounds identified as being used in hydraulic fracturing
to extract gas, 65 are listed as hazardous by the federal
government.3 In turn, this creates a real potential for negative Drilling directly behind the play yard.
health effects in any area where gas development occurs.4
Photo by: Nadia Steinzor
While general scientific links regarding the effects of such
exposure have been established,5 research on the direct
relationship between health problems and gas and oil
activities has been limited and inconsistent.6

Even as knowledge of impacts evolves slowly, gas and oil extraction and production continue to
accelerate rapidly—allowing industry to put still-emerging technologies to use without first
establishing their safety. State regulations remain too lax and outdated to prevent the impacts of
modern-day energy development, and regulatory agencies are often unable to conduct the


oversight and enforcement needed to protect air and water quality and, in turn, health and
communities. Magnifying the consequences of this situation are special exemptions in provisions
of the nation’s bedrock environmental laws, which allow the industry to stifle key information and
pursue risky practices.7

The overall result is that the burden of proof remains heaviest for impacted individuals and
communities themselves. Companies can continue to avoid responsibility and downplay health-
related concerns. Decisionmakers can continue to sidestep the need to recognize the damage and
hold companies accountable.

Yet the realities, including those described in this report, can be documented—and when they are,
they can no longer be denied. When many people in many places where gas development is
occurring have similar health complaints, something is clearly wrong. OGAP believes that when
health problems occur, action to solve and prevent them must follow.


1 Background: The Marcellus Shale Boom
The Marcellus Shale spans a distance of approximately 600 miles from central New York through
much of Pennsylvania and into the eastern half of Ohio and western parts of West Virginia; small
sections are also found in Maryland and Virginia.8 The formation covers an area of about 54,000
square miles (slightly larger than Florida) and varies greatly in depth, from outcroppings above
ground to some areas to 9,000 or more feet below the surface.9

For a long time, extracting and producing deep shale gas from formations across the United States
was considered economically and technologically unfeasible. But recent advances in hydraulic
fracturing methods and its combination with horizontal drilling have made it possible to drill much
deeper and further than ever before and, bolstered by political pressure to expand domestic
energy production, have spurred a boom in shale gas (and shale oil) production nationwide.

The Marcellus Shale, considered a “gas super
giant,” has been at the center of this activity,
particularly in Pennsylvania, an estimated 60
percent of which is underlain by the formation.10
As of September 2012, nearly 5,900
unconventional oil and gas wells, primarily in the
Marcellus Shale, had been drilled in the state and
over 11,500 had been permitted; the pace of
expansion has been stunning, with 75 percent of
all unconventional wells having been drilled just
since 2010.11
Rigs in the neighborhood.
Gas and oil development is occurring in
Pennsylvania and nearby states today more Photo by: Mark Schmerling

rapidly and with more extensive impacts than in
the past. Current development uses a
tremendous amount of water, chemicals, and
land; requires heavy equipment; and produces
large volumes of both wastewater and solid
waste. The gas industry has plans for tens of
thousands of additional wells across the Marcellus
and Utica Shale regions and in other formations
nationwide.

The complexity and intensity of this type of
energy development opens up pathways of
exposure that impact human health, including air
Impoundment pits, which often contain contaminated waste,
and water pollution, traffic, noise, and soil can leak and give off emissions.
contamination. Although no industrial process is
Photo by: Robert Donnan
harm-free, these problems can be particularly
severe when operators act irresponsibly and are
not required to take measures to prevent,
minimize, or mitigate problems such as chemical
and waste spills, emission releases, or equipment
failure.


2 The Study
2.1. OVERVIEW
This research project had two central components, health surveys and air and water testing, and
was undertaken in order to:

 Investigate the extent and types of health symptoms experienced by people living in the
gas patches of Pennsylvania.
 Consider links between health symptoms and proximity to gas extraction and production
facilities.
 Provide air and water quality testing to households in need of such information.
 Provide useful information to impacted residents, researchers, public officials, and partner
organizations.
 Put forth common-sense recommendations for
regulatory and policy changes to prevent
negative health and environmental impacts.

This project represents a scaling-up (in terms of both the
number of participants and geographic area covered) of
community-based projects previously conducted by
Earthworks’ OGAP. We conducted health surveys with
local residents and analyzed results in relation to
contaminants identified through water quality
investigations (Pavillion, Wyoming, 2010) and prior air
quality monitoring (DISH, Texas, 2009).12 In addition, in
2011 OGAP developed case studies of residents who Separators split off heavy hydrocarbons from gas, and
reported health problems while living in close proximity often vent methane and volatile organic compounds into
to gas facilities in several counties in Texas.13 the air.
Photo by: Nadia Steinzor

2.2. METHODOLOGY
The health survey instrument used in this project was designed by Wilma Subra, President of Subra
Company, and air and water quality testing was managed by the non-profit organization ShaleTest
based in Denton, Texas. Data from the surveys and associations with testing results were obtained
by tabulating responses and calculating percentages of both symptom categories and individual
symptoms.

The survey focused on a range of exposures, health symptoms, and disease history. Responses
were gathered to identify patterns that occur across locations and improve understanding of the
experiences of participants. All the symptoms included in the survey could potentially be caused
by exposure to substances known to be associated with gas and oil facilities.

It should be noted that this project did not investigate additional factors that can influence health
conditions or cause symptoms (e.g., through structured control groups in non-impacted areas and
in-depth comparative health history research). Such work, while important and currently lacking,
was beyond the scope of this particular project. In addition, we did not seek to link single facilities
with particular health problems experienced by specific participants.


The survey was completed by 108 individuals living in 55
households in 14 counties. The largest number of surveys
(85 percent) was collected in Bedford, Bradford, Butler,
Fayette, and Washington Counties. Taken together, all the
counties represent a geographical range across the state
(i.e., northeast to southwest) and have had gas
development long enough for reports of health impacts
and declining air and water quality to surface.

Respondents answered questions on their own or
provided them to a relative or friend. In some cases,
members of the same household, including spouses and
parents, completed surveys for participants, and a few
participants chose to provide answers to OGAP staff in
person or over the phone. Due to expressed concerns Condensate tanks pull off water from gas
about confidentiality, participants were given the option and can produce fumes and emissions.
of completing the surveys anonymously, which some Photo by: Nadia Steinzor
chose to do.14

Survey distribution was initiated through existing contacts
in the target counties. These individuals then chose to
participate in the project themselves and/or
recommended other possible participants, who in turn
provided additional contacts. The survey was also
distributed to individuals who expressed interest in
participating directly to OGAP at public events.

Air and water are the primary pathways of exposure to
chemicals and other harmful substances, which are
inhaled, ingested, and absorbed through the skin. With
this in mind, environmental testing was conducted on the Placing an air canister at a test site for collecting
properties of a subset of survey participants (70 people in emissions data.
total) in order to identify the presence of pollutants that Photo by: Nadia Steinzor
might be linked to both gas development and health
symptoms. Test locations were selected based on
household interest, the severity of symptoms reported, and proximity to gas facilities and
activities. Because the need for testing in such places far exceeded the resources available, we also
considered whether households had already received other environmental testing and been
provided with the results.

In total, 34 air tests and 9 water tests were conducted at 35 households in 9 counties. The air tests
were conducted using Summa Canisters put out for 24-hours by trained members of ShaleTest.
The samples were analyzed by three certified laboratories using U.S. Environmental Protection
Agency-approved TO-14 and TO-15 methods, which test for a wide range of Volatile Organic
Compounds (VOCs) such as benzene, toluene, ethylbenzene, and xylene (BTEX chemicals). The
water tests used samples drawn directly from household sinks or water wells by technicians
employed by licensed laboratories and covered the standard Tier 1, Tier 2, and Tier 3 (including
VOCs/BTEX) and in one case, Gross Alpha/Beta, Radon, and Radium as well.


2.3. FINDINGS

PARTICIPANT OVERVIEW
Among participants, 45 percent were male from 18 months to 79 years of age and 55 percent were
female from 7 to 77 years of age. The closest a participant lived to gas facilities was 350 feet and
the furthest away was 5 miles.

Participants had a wide range of occupational backgrounds, including animal breeding and
training, beautician, child care, construction, domestic work, farming, management, mechanic,
medical professional, office work, painter, retail, teaching, and welding. About 20 percent of
participants reported occupational-related chemical exposure (for example, to cleaning products,
fertilizers, pesticides, and solvents). At the time of survey completion, 80 percent of participants
did not smoke and 20 percent did. While some of the non-smokers had smoked in the past, more
than 60 percent never had.

Table 1: Survey location

County surveyed Number of surveys collected Percent of surveys
Washington 24 22
Fayette 20 18
Bedford 20 18
Bradford 17 16
Butler 12 11
Jefferson 3 3
Sullivan 2 2
Greene 2 2
Warren 2 2
Elk 2 2
Clearfield 1 1
Erie 1 1
Susquehanna 1 1
Westmoreland 1 1
TOTAL 108 100%


A COMPARATIVE LOOK:
We established an informal comparison group of 5 individuals in 5 households in and around the
city of Sayre in north-central Bradford County and in Waverly, New York, just over the state border
from Sayre. This group generally lived further away from gas facilities than the main survey group,
between 1.5 - 9 miles from gas wells and (in one case) 8 miles from a compressor station.
None of the participants smoked and all reported being healthy. Taken together, these
participants reported a total of 24 health symptoms, including some in the categories of skin,
respiratory, digestive, muscles/joints, neurological, ear/nose/mouth, behavioral, and lymphatic.
Only one or two participants reported each symptom or smelling odors of any kind—reflecting a
lower level of impact than was generally documented among survey participants overall. While
much smaller than the main survey group, the comparison group results indicate the possibility
that fewer health symptoms exist at longer distances from facilities, an aspect indicated by the
project findings overall that warrants further investigation and analysis.

HEALTH SYMPTOMS
Almost half of the survey participants answered the question of whether they had any health
problems prior to shale gas development. About half of those responses indicated no health
conditions before the development began and about half reported
having had one or just a few, in particular allergies, asthma, arthritis,
cancer, high blood pressure, and heart, kidney, pulmonary, and 22 households reported that pets
thyroid conditions. and livestock began to have
symptoms (such as seizures or
In addition, 5 individuals volunteered (verbally or in writing) that losing hair) or suddenly fell ill
their existing health symptoms became worse after shale gas and died after gas development
development started and 15 that their symptoms lessened or began nearby.
disappeared when they were away from home. Members of four
households also reported that they’d moved to new locations due to
gas drilling and several others told OGAP staff that they would if their
finances and jobs allowed it. Also of note is that participants in 22
households reported that pets and livestock began to have unexplained
symptoms (such as seizures or losing hair) or suddenly fell ill and died
after gas development began nearby.

The specific symptoms reported within each of the top reported
categories varied.15 (To see which specific symptoms were included in all
the categories, see the full survey at http://health.earthworksaction.org.)
However, the primary categories of health problems reported by Industrial equipment containing
participants were quite consistent across counties. For example, combustible products, very close to home.
sinus/respiratory problems was the top complaint category for all Photo by: Nadia Steinzor
participants, as well as in four of the five main counties and the other
counties group; the second top complaint category,
behavioral/mood/energy, was the first in one county, second in three
and in the other counties group, and third and fourth in one each.


Table 2: Ranking of top 8 categories of symptoms, by county
Symptom Category All Bedford Bradford Butler Fayette Washington Others*
Sinus/respiratory 1 1 2 1 1 1 1
Behavioral/mood/energy 2 2 1 3 2 4 2
Neurological 3 5 5 8 7 3 3
Muscles/joints 4 3 3 4 8 5 6
Ear/nose/mouth 5 7 8 7 5 6 5
Digestive/stomach 6 4 7 5 4 8 7
Skin reactions 7 6 4 2 3 7 8
Vision/eyes 8 8 6 6 6 2 4

Table 3: Most prevalent categories of symptoms
Percent of individuals reporting conditions in category
Symptom category All Bedford Bradford Butler Fayette Washington Others*
Behavioral/mood/energy 80 60 88 67 85 74 67
Neurological 74 45 71 50 70 79 60
Muscles/joints 70 55 82 67 70 74 47
Digestive/stomach 64 55 65 58 75 63 33
Ear/nose/mouth 66 40 59 50 75 68 47
Skin reactions 64 45 70 67 75 63 27
Vision/eyes 63 40 65 50 70 79 53
* Includes Clearfield, Elk, Erie, Jefferson, Greene, Sullivan, Susquehanna, Warren, and Westmoreland Counties. The
surveys from these counties (15) were analyzed together to create a group comparable in number to each of the
counties where more surveys were collected.

The 25 most prevalent symptoms among all participants were increased fatigue (62%), nasal
irritation (61%), throat irritation (60%), sinus problems (58%), eyes burning (53%), shortness of
breath (52%), joint pain (52%), feeling weak and tired (52%), severe headaches (51%), sleep
disturbance (51%), lumbar pain (49%), forgetfulness (48%), muscle aches and pains (44%), difficulty
breathing (41%), sleep disorders (41%), frequent irritation (39%), weakness (39%), frequent nausea
(39%), skin irritation (38%), skin rashes (37%); depression (37%), memory problems (36%), severe
anxiety (35%), tension (35%), and dizziness (34%).

The survey asked questions designed to identify if there might be associations between symptoms
and living near particular types of facilities (wells, waste impoundment pits, and compressor
stations). However, because it turned out that most survey participants actually live in close
proximity to more than one type of facility, it was difficult to determine connections with a specific
type of facility. Instead, we examined whether the distance from any type of oil and gas facility had
a bearing on the number of types of symptoms reported in the survey.


As seen in Table 4, many symptoms were commonly reported regardless of the distance from the
facility (in particular sinus problems, nasal irritation, increased fatigue, feeling weak and tired, joint
pain, and shortness of breath).

In general, as the distance from facilities decreases, the percentage of respondents reporting the
symptoms increases. For example, when facilities were 1500-4000 feet away, 27 percent reported
throat irritation; this increased to 63 percent at 501-1500 feet, and 74 percent at less than 500 feet.
For severe headaches, 30 percent reported them at the longer distance, but about 60 percent at
the middle and short distances.

However, when facilities were further away than 4001 In general, the closer to gas facilities
feet, some percentages jumped back up. The data respondents lived, the higher the rates
showed higher percentages of respondents of symptoms they reported.
experiencing certain symptoms at the longer than mid-
range distances with regard to several other symptoms
(e.g., throat irritation, sinus problems, nasal irritation, eye
burning, and joint pain). It is possible that the chemicals
that bring on these types of symptoms travel over much
longer distances than would normally be expected, or
that other factors were at play related to the landscape,
weather conditions, participant reporting, and type of
production.16

When the most prevalent symptoms are broken out by
age and distance from facility, some differences are
notable. In most age groups, symptoms are more
prevalent in those living closer to facilities than those
living further away. In sum, while the data presented in
Figure 1 below do not prove that living closer to an oil
and gas facility causes health problems, they do suggest
a strong association.

Drilling rig onsite.
Photo by: Frank Finan


Table 4. Differences in symptoms by distance from an oil/gas facility
Distance Number Age Number Average Number of Top 15 symptoms
of surveys range of number of symptoms (the / means the same percentage of
in this smokers symptoms experienced respondents had the symptom – bringing
category per person by 50% or the total to more than 15 symptoms in
more some cases)
respondents

Within 500 27 1.5 to 2 31 9 Throat irritation (74%), sinus problems
feet of any 76 (1 (70%), nasal irritation/ eye burning/ joint
facility had no pain/ severe headaches/ sleep
age disturbances (59%), skin rashes (56%),
data) shortness of breath (52%), loss of sense
of smell/ persistent cough/forgetfulness/
sleep disorders/ frequent nosebleeds/
swollen painful joints/ increased fatigue/
feeling weak and tired (44%)
501 - 1500 40 3 to 79 12 30 11 Increased fatigue (68%), nasal
feet from a irritation(65%), throat irritation (63%),
facility eye burning/ severe headaches (60%),
shortness of breath (55%), sleep
disturbances/ sinus problems/ lumbar
pain (53%), feeling weak and tired/
forgetfulness (50%), joint pain/ muscular
pain/ memory problems/ weakness
(48%)
1501 - 30 6 to 77 9 27 1 Increased fatigue (57%), feeling weak
4000 feet (1 had and tired (47%), joint pain (43%),
from a no age shortness of breath/ difficulty breathing
facility data) (40%), sinus problems/ lumbar pain/
forgetfulness/ tension/ weakness of
hands (37%), nasal irritation/ frequent
nausea/ reduced muscles strength/
persistent skin problems (33%)
Greater 11 34 to 2 29 8 Throat irritation/ nasal irritation/ feeling
than 4001 76 weak and tired (64%), sinus problems/
feet from a eye burning/ joint pain/ muscle aches or
facility pains/ ringing in ears (55%), increased
fatigue/ severe headaches/ shortness of
breath/ sleep disturbances/ lumbar pain/
muscular pain/ weakness/ depression/
persistent hoarseness/ blurred vision
(45%)


Figure 1. Association of symptoms and distance from facilities, by age group
In the youngest age group
(1.5-16 years old), the most
common symptoms were
related to sensitive mucous
membranes (throat, eyes,
nose) and skin. Even these
youngest respondents had
conditions not typically
associated with children
(e.g., severe headaches,
joint pain, lumbar pain,
and forgetfulness).17
In the subset of this young
age group living 1500 feet or
closer to a facility, the
percentage of respondents
with symptoms increased.
For example, the number of
respondents experiencing
throat irritation jumped from
57 to 69 percent, and severe
headaches increased from
52 to 69 percent. Of all age
groups, this group had the
highest occurrence of
frequent nosebleeds
within 1500 feet of
facilities (56%).18

In the next age group (20-40
years old), there was a high
occurrence of symptoms
related to the throat, eye,
and nose; fatigue, nausea,
and severe headaches were
also common symptoms.
For those living 1500 feet or
closer to a facility, the
percentage of respondents
with symptoms increased for
all symptoms except one
(headaches). In some cases,
the percentage reporting
symptoms was considerably
higher (e.g., for sinus
problems, eye burning,
shortness of breath, and
sleep disturbances). 44
percent of 20 to 40-year-
olds living within 1500 feet
of facilities complained of
frequent nosebleeds,
compared to 29 percent of
all participants of this age.


Approximately 60 percent of
participants in the third age
group (41-55 years old)
reported throat irritation,
increased fatigue, nasal
irritation, joint pain, and
severe headaches.
Although the occurrence of
some symptoms (e.g., throat
irritation, joint pain, and
sleep disturbances)
increased in the subset of
this group living closer to
facilities, the increases were
not as dramatic as those
experienced in other age
groups. In some cases, the
percentages actually went
down in the subgroup of
those living closer to
facilities.

In the oldest age group (56-
79 years old) the symptoms
most frequently experienced
were increased fatigue,
shortness of breath, and
feeling weak and tired.
For some symptoms (e.g.,
throat irritation, sinus
problems, nasal irritation,
eye burning, shortness of
breath, severe headaches
and skin rashes) there were
large increases in the
subset of this age group
who lived closest to
facilities.


The survey also asked respondents to indicate whether or not they were smokers. Table 5 shows
that while smokers had, on average, more symptoms than non-smokers, and more symptoms in
common with each other, the most frequently reported symptoms were very similar to non-
smokers (including forgetfulness, increased fatigue, lumbar pain, joint pain, eye burning, nasal
irritation, sinus problems, sleep disturbances, severe headaches, throat irritation, shortness of
breath, frequent nausea, muscle aches or pains, and weakness).

The fact that the non-smokers reported symptoms that are commonly considered to be side
effects of smoking (e.g., persistent hoarseness, throat irritation, sinus problems, nasal irritation,
shortness of breath, and sleep disturbances) suggests that there are likely factors other than
smoking that contribute to these symptoms.

Table 5. Comparison of symptoms in smoking and non-smoking subgroups of similar ages 19
Number Age Average Number of Top 15 symptoms (in order of highest
in this range number of symptoms percentage reporting symptom)
category symptoms experienced
per by 50% or
person more of
respondents
Non- 54 23 - 70 27 6 Forgetfulness (59%), lumbar pain/ joint pain
smokers (57%), increased fatigue (56%), eye burning/ nasal
irritation (54%), sinus problems/ sleep
disturbances (48%), severe headaches/ throat
irritation (44%), shortness of breath (43%),
frequent nausea/ muscular pain/ persistent
hoarseness (41%), weakness (39%)

Smokers 27 24 - 70 38 13 Increased fatigue (70%), eye burning/ lumbar pain
(59%), sinus problems/ nasal irritation/ joint pain/
forgetfulness/ severe headaches/ sleep
disturbances (56%), shortness of breath/ throat
irritation/ frequent nausea/ muscular pain (52%),
feeling weak and tired/ weakness (48%)

Breaking down the data further, as shown in Table 6, it appears that the symptoms most
frequently reported by smokers and non-smokers were remarkably similar within each age
group. For example, in the age group 20-40, increased fatigue, sinus problems, throat irritation,
frequent nausea, and sleep problems were among the top symptoms for smokers and non-
smokers. In the 41-55-year-old group, increased fatigue, throat irritation, eye burning, severe
headaches, and feeling weak and tired were among the top symptoms in both groups, and in the
over-56 age group, eye burning, sinus problems, increased fatigue, joint pain, and forgetfulness
were among the top symptoms of smokers and non-smokers.

Furthermore, the data from smokers did not greatly affect the results in the “all respondents”
category. When compared to the non-smoking subgroup, the only notable difference was in the
41-55-year-old age group, where the average number of symptoms in the “all respondents” was
30, versus 22 in the non-smoking subgroup. The top symptoms, however, were very similar.


Table 6. Symptoms by age group, and by smoking or non-smoking status
Age Sub-category Number Average Number of Top 15 symptoms (in order of highest percentage
category in sub- number of symptoms in reporting the symptom)
category symptoms 50% or more
per person respondents

16 and All respondents 21 19 2 Throat irritation (57%), severe headaches
under (None were (52%), nasal irritation (48%), skin rashes/
smokers) abdominal pain/ eye burning/ frequent nose
bleeds/ sleep disturbances (43%), sinus
problems/ persistent cough (38%), shortness
of breath/ frequent nausea (33%), skin
irritation/ asthma/ difficulty breathing/
allergies/ diarrhea/ dry eyes/ muscle aches or
pains/ forgetfulness/ behavioral changes/
frequent irritation (29%)
20 - 40 All respondents 14 29 12 Increased fatigue (64%), severe headaches/
sinus problems/ throat irritation/ frequent
nausea (57%) abdominal pain/ nasal irritation/
eye burning/ muscular pain/ lumbar pain/
weakness/ sleep disturbances/ depression
(50%), dry/cracked red skin/ feeling weak and
tired/ sleep disorders/ allergies/ sores or ulcers
in mouth/ forgetfulness/ joint pain/ severe
anxiety (43%)
Non-smokers 10 29 10 Increased fatigue/ severe headaches/
abdominal pain, (60%), sinus problems/ throat
irritation/ frequent nausea/ nasal irritation/
dry, cracked red skin/ feeling weak and tired/
sleep disorders (50%), eye burning/ muscular
pain/ lumbar pain/ sleep disturbances/
depression/ allergies/ sores or ulcers in mouth/
forgetfulness/ skin rashes/ shortness of breath/
diarrhea/ extreme drowsiness/ tension/
persistent skin problems/ loss of sense of
smell/ lumps or swelling neck (40%)
Smokers 4 28 28 Weakness (100%), increased fatigue (75%),
sinus problems (75%), throat irritation (75%),
frequent nausea (75%), eye burning (75%),
muscular pain (75%), lumbar pain (75%), sleep
disturbances (75%), depression (75%), joint
pain (75%), severe anxiety (75%), frequent
irritation (75%), severe headaches/ nasal
irritation/ allergies/ sores or ulcers in mouth/
forgetfulness/ swollen painful joints/ muscle
aches or pains/ loss of sex drive/ irregular or
rapid heart beat/ persistent hoarseness/
reduced muscle strength/ difficulty
concentrating/ severe pain in eyes/
compulsive behavior/ weight loss (50%)


41 - 55 All respondents 33 30 8 Severe headaches/ nasal irritation/ increased
fatigue (63%), joint pain/ throat irritation
(60%), feeling weak and tired (57%), sinus
problems/ eye burning (60%), shortness of
breath/ sleep disturbances/ depression,
muscles aches or pains (49%), lumbar
pain/forgetfulness (46%), memory problems
(43%)
Non-smokers 22 22 8 Joint pain (68%), nasal irritation (64%), throat
irritation (59%), severe headaches/ increased
fatigue/ feeling weak and tired (55%), sinus
problems/ depression (50%), eye burning/
muscle aches or pains (45%), shortness of
breath/ memory problems/ lumbar pain/ skin
rashes (41%)
Smokers 13 44 18 Severe headaches/increased fatigue (77%),
sleep disturbances (69%), nasal irritation/
throat irritation / feeling weak and tired/ eye
burning/ shortness of breath/ forgetfulness/
sleep disorders/ loss of sex drive (62%), sinus
problems/ muscle aches or pains/ lumbar
pain/ skin irritation/ muscular pain/ persistent
hoarseness/ agitation (54%)
56 - 79 All respondents 36 32 11 Increased fatigue (67%), shortness of breath/
feeling weak and tired (64%), sinus problems/
eye burning, joint pain (56%), forgetfulness
(53%), difficulty breathing/ nasal irritation/
lumbar pain/ sleep disturbances (50%), throat
irritation (47%), weakness/ reduced muscle
strength/ memory problems (44%)
Non-smokers 28 32 9 Feeling weak and tired (68%), increased
fatigue/ shortness of breath (64%), sinus
problems/ eye burning/ sleep disturbances
(54%), joint pain/ forgetfulness/ throat
irritation (50%), difficulty breathing/ nasal
irritation/ weakness/ memory problems/ sleep
disorders/ frequent urination (46%)
Smokers 8 35 18 Increased fatigue/ joint pain/ lumbar pain
(75%), shortness of breath/ sinus problems/
eye burning, forgetfulness/ difficulty
breathing/ nasal irritation/ tension/ frequent
nausea (63%), feeling weak and tired/ reduced
muscle strength/ arthritis/ muscular pain/
persistent skin problems/ diarrhea/ skin rashes
50%)


ODOR EVENTS
Bad and unusual odors are an indication of the presence of a
substance or chemical, and are a common complaint of people
living near gas facilities. Among survey participants, 81 percent 81% reported experiencing
bad odors sometimes or constantly.
reported experiencing bad odors sometimes or constantly. The
frequency ranged from 1 to 7 days per week and from several The frequency ranged from 1 to 7 days
times per day to all day long; 18 percent said they could smell per week and from several times per day
odors every day. to all day long

Participants were asked to describe the suspected source of the
odors. Nearly all responses related odors to gas facilities and
events, including drilling; gas wells; well pads; fracturing;
compressor stations; condensate tanks; drinking contaminated
water; flaring; waste pits; retention ponds; diesel engines; truck
traffic; pipelines and pipeline stations; spills and leaks; subsurface
events; seismic testing; and blue-colored particles in air (possibly a
sign of catalytic compounds or particulate matter).

When asked in the survey whether health symptoms occurred in
conjunction with odor events, participants reported the
associations listed below. Most indicated that symptoms would
last from a few hours to a few days and, in some cases, a few
Centralized compressor stations move large volumes
weeks.
of gas to and through pipelines. Emissions can
include volatile organic compounds such as benzene
 Nausea: ammonia, chlorine, gas, propane, ozone, rotten and toluene, nitrogen oxides, and formaldehyde.
gas. Photo by: Nadia Steinzor
 Dizziness: chemical burning, chlorine, diesel, ozone,
petrochemical smell, rotten/sour gas, sulfur.
 Headache: chemical smell, chlorine, diesel, gasoline,
ozone, petrochemical smell, propane, rotten/sour gas,
sweet smell.
 Eye/vision problems: chemical burning, chlorine, exhaust.
 Respiratory problems: ammonia, chemical burning, chlorine, diesel, perfume smell, rotten
gas, sulfur.
 Nose/throat problems: chemical smell, chlorine, exhaust, gas, ozone, petrochemical smell,
rotten gas, sulfur, sweet smell.
 Nosebleeds: kerosene, petrochemical smell, propane, sour gas.
 Skin irritation: chemical smell, chlorine, ozone, sulfur.
 Decreased energy/alertness: chemical gas, ozone, rotten/sour gas, sweet smell.
 Metallic/bad taste in mouth: chemical burning, chlorine, turpentine.


2.4. ENVIRONMENTAL TESTING

AIR
As seen in Table 7, the 34 Summa canister air tests, taken together, detected a total of 19 VOCs. In
sum, there was considerable consistency in the chemicals present in many of the samples,
although concentrations varied. This could in part be due to differences in the reporting limits and
suite of chemicals analyzed by the three labs used in this project. It is possible, for example, that
more VOCs were present in more locations, but Pace Analytical had much higher reporting limits
than Columbia and Con-Test so the Pace results showed “non-detect” for many substances.20

Table 7. VOCs in ambient air, sorted by highest percent detection; concentrations are in micrograms per
cubic meter, μg/m3 (n = total number of canister samples that were analyzed for a particular chemical; NA =
VOC not included in the analysis)
Volatile Organic Compound n Number of Percent of Min. Max. Mean* Chemical reporting limits for the
(VOC) samples n three labs used
detecting detecting
Columbia Con- Pace**
VOC VOC
Test

2-Butanone 17 16 94 0.95 2.9 1.52 0.85 - 1.3 NA NA
Acetone 17 15 88 8.0 19 11.85 6.5 - 10 NA NA
Chloromethane 34 27 79 1.0 1.66 1.21 0.59 - 0.90 0.1 1.39 1.53
1,1,2-Trichloro-1,2,2- 34 26 76 0.54 0.73 0.64 0.22 - 0.34 0.38 5.13 - 5.67
trifluoroethane
Carbon tetrachloride 34 26 76 0.46 0.76 0.62 0.091 - 0.31 4.21 - 4.65
Trichlorofluoromethane 34 26 76 0.6 1.8 1.48 0.81 -1.2 0.28 3.32 - 3.66
Toluene 34 22 65 0.68 7.9 1.83 0.53 – 0.82 0.19 2.52 - 2.79
Dichlorodifluoromethane 17 9 53 1.9 2.8 2.41 NA 0.25 3.32 - 3.66
n-Hexane 8 3 38 3.03 7.04 5.23 NA NA 2.37 - 2.61
Benzene 34 11 32 0.31 1.5 0.85 0.46 - 0.67 0.16 2.14 - 2.36
Methylene Chloride 34 10 29 1.9 32.62 7.93 0.49 – 0.76 1.7 2.33 - 2.57
Total Hydrocarbons (gas) *** 8 2 25 49.8 146 97.9 NA NA 46.9 - 52.2
Tetrachloroethylene 34 8 24 0.12 10.85 1.68 0.10 – 0.16 0.34 4.54 - 5.02
1,2,4-Trimethylbenzene 17 4 24 0.38 0.61 0.48 NA 0.25 3.30 - 3.64
Ethylbenzene 34 6 18 0.27 1.5 0.54 1.4 – 1.9 0.22 2.91 - 3.21
Trichloroethylene 34 6 18 0.17 5.37 2.71 0.08 - 0.12 0.27 3.60 - 3.98
Xylene (m&p) 34 5 15 0.92 5.2 1.98 2.5 – 3.8 0.43 2.82 - 3.12
Xylene (o) 34 5 15 0.39 1.9 0.76 1.2 – 1.9 0.22 2.91 - 3.21
1,2-Dichloroethane 34 1 3 0.64 0.64 0.64 0.59 - 0.90 0.2 2.71 - 2.99

* Mean of samples detecting chemical. 21
** Pace reporting limits were in ppbv. We converted to μg/m3.22
*** Total hydrocarbons reported as parts per billion volume (ppbv).


Breaking out the air data by county, the highest number of VOCs were detected in samples from
Washington County (15), Butler County (15), Bradford County (12), and Fayette County (9).
Washington County also had the highest measured concentration of five and the second highest
concentration of 12 VOCs.23 Samples from Butler and Bradford Counties had the highest
concentrations of five and three VOCs, respectively. Five chemicals were detected in all nine of the
samples from Washington County and in the six samples from Butler County: 1,1,2-Trichloro-1,2,2-
trifluoroethane; carbon tetrachloride; chloromethane; toluene; and trichlorofluoromethane.
(Detailed data for all the counties where air testing occurred are available at
http://health.earthworksaction.org.)

In 2010, the Pennsylvania Department of Environmental Protection (DEP) conducted air testing
around natural gas wells and facilities in three regions across the state, in part using the same
canister sampling methods as in this project.24 When compared to the DEP’s results, OGAP’s results
showed some similarities in both the chemicals detected and concentrations.

Figure 2 shows benzene, toluene, ethylbenzene, and m&p-xylenes (o-xylenes not included) broken
down by county from our project, as well as samples taken by DEP at control sites (rural, forested
areas with no nearby gas development), oil and gas sites (including some nearby residences), and
an industrial site (the Marcus Hook monitoring site, which is close to two oil refineries in an
industrialized area of the state25). Also shown in the chart are the number of detections and the
number of samples in each category (e.g., benzene was detected in four of six air samples in Butler
county).

As seen in these charts, BTEX chemicals measured in our project in Butler and Washington
Counties were consistently higher than concentrations found at DEP control sites (ethylbenzene
and m&p-xylenes were not detected at any of the control sites). When compared to the sampling
done by DEP around oil and gas facilities the concentrations in Butler
and Washington Counties were in the same range for benzene, but
were considerably higher for toluene, ethylbenzene, and m&p-xylenes. Concentrations of
It is also striking that some of the concentrations of ethylbenzene and ethylbenzene and xylene
xylene measured at homes in Butler and Washington Counties were measured at homes in
higher than any concentration detected by the DEP at the Marcus Hook Butler and Washington
industrial site. Again, while factors such as topography, type of gas, and Counties for this project
emission control technologies can influence air results, it is highly were higher than in air
possible that air quality at the sites where we tested—all in rural and tests done by the DEP at
residential areas—was worse overall because of the proximity of gas the Marcus Hook
facilities. Industrial site.
According to the DEP, some of the VOCs found in our study are present
in ambient air because they were once widely used and persist in the
atmosphere.26 The DEP indicates that acetone and the BTEX chemicals, however, may be attributed
to gas development.27 (In addition, the presence of VOCs clearly influences air quality overall.)


Figure 2. The four charts that make up Figure 2 show comparisons of BTEX concentrations where they
were detected in project samples (Bradford, Butler, Fayette, and Washington Counties only) and DEP
samples (control, oil and gas facility, and an industrial site).


To provide some perspective on benzene concentrations found in our results, we examined data
on national benzene concentrations in the U.S. (based on annual average concentrations at 22
urban sampling locations). Between 1994 and 2009, benzene in ambient air declined.28

In 2009, 80 percent of the urban sites had average annual benzene concentrations between 0.4
and 1.5 micrograms per cubic meter (μg /m3 ), with the average and median concentrations for the
22 sites being less than 1 μg /m3. Five of our air canister tests had benzene above the national
(urban) average, and two had concentrations equal to the maximum average annual concentration
measured by EPA in U.S. urban areas in 2009 (i.e., 1.5 μg /m3).29


As mentioned previously, the current project’s sampling locations were in
rural areas of Pennsylvania. While local traffic may have contributed some
benzene, the most likely primary sources of benzene in these areas are oil
and gas facilities; increased truck traffic associated with these sites could
also be a contributing factor.

It is important to note that the concentrations found in our study were one-
time samples, while the EPA concentrations represent an average of many
samples taken over the course of a year. So there may have been some
individual samples in urban areas that were higher than 1.5 μg /m3. It is also Operators flare gas that's uneconomical
to process or to burn off certain
possible, however, that benzene concentrations at the sampling locations compounds. Flaring emits a host of air
in our project could have exceeded 1.5 μg /m3 if numerous samples were pollutants determined by the chemical
taken over the course of several months or a year. composition of the gas and the
temperature of the flare.
Finally, the chemicals sampled in our project were limited to a selection of Photo by: Frank Finan
VOCs. The analytical methods used did not test for some chemicals known
to be associated with oil and gas facilities such as formaldehyde, which is
commonly emitted from compressor stations. According to the U.S. EPA,
the major toxic effects caused by acute formaldehyde exposure via
inhalation are eye, nose, and throat irritation and effects on the nasal
cavity.30 These were symptoms experienced by high percentages of survey
respondents. In addition, hydrogen sulfide, a known toxic compound with
many of the health effects documented in this project, is often associated
with oil and gas development. Testing for such chemicals would have
required different types of air sampling methods than applied here.

WATER
The nine water samples taken for this project were sent to laboratories
that analyzed for dozens of substances. Table 8 shows the 26 Drinking water standards do
parameters that were detected in at least one sample (including water not even exist for some
temperature and pH). Several of the chemicals found in samples are
contaminants, such as methane,
known to be associated with oil and gas drilling operations. For
bromide, sodium, strontium, or
example, barium, bromide, calcium, chloride, iron, manganese,
Total Suspended Solids (TSS).
magnesium, potassium, sodium, sulfate, strontium, and Total Dissolved
Solids (TDS) have been measured in effluent from a Pennsylvania
wastewater plant that only treats oil and gas industry brine and
hydraulic fracturing flowback.31


Table 8. Water quality results from nine private water wells in Bradford and Butler Counties, Pennsylvania
(Note: not all parameters were analyzed in every sample)
Parameter Units Number Number Min.a Max. Meanb PA Number of
of above DEP samples
samples detection MCLc above MCL
limit

Barium mg/L 9 9 0.029 0.5 0.25 2 0
Calcium mg/L 9 9 33 66.2 43.7 None
Magnesium mg/L 9 9 4.5 16.8 9.1 None
Sodium mg/L 9 9 9.2 64.1 20.9 None
Strontium mg/L 9 9 0.126 1.7 0.5 None
Hardness (Total as mg/L 9 9 120 234 147 None
CaCO3)
pH Std Units 9 9 6 7.9 6.5 6.5 - 8.5 2 below
Alkalinity (Total as mg/L 9 9 38 285 130 None
CaCO3)
Total Dissolved Solids mg/L 9 9 138 392 218 500 0
Sulfate mg/L 9 9 6.7 231 33 250 0
Manganese mg/L 9 7 <0.005 6.44 1.04 0.05 7
Chloride mg/L 9 7 <5.0 84.3 24.1 250 0
Iron mg/L 9 6 <0.04 153 19.5 0.3 5
Potassium mg/L 6 6 1.14 1.57 1.1 None
Specific Conductance μmhos/cm 6 6 287 552 326 None
Methane μg/L 9 5 1.06 57.4 10 None
Arsenic mg/L 9 4 <0.001 0.0282 0.005 0.010 1
Lead mg/L 9 4 <0.001 0.113 0.01 0.005* 3
Total Coliform per 100 mL 9 4 Absent Present None
Total Suspended mg/L 6 4 <5 448 118 None
Solids
Temp, water Deg. Celsius 3 3 25 29 28 None
Turbidity NTU 3 3 0.22 5.7 2.3 None
Nitrate mg/L 3 3 0.076 0.71 0.46 10 0
E. coli per 100 mL 9 2 Absent Present None
Sulfur μg/L 1 1 <1,000 7,550 2,850 None
Bromide mg/L 1 1 0.26 0.26 0.26 None
a Minimum values: If reports included non-detects of a particular chemical, the minimum value in the table was shown as being less than
(<) the lowest laboratory detection limit.
b Mean values: Non-detected chemicals were assigned a concentration equal to half of the detection limit only if there were other samples

that detected the chemical.
c MCL: Maximum Contaminant Levels published by the Pennsylvania Department of Environmental Protection Division of Drinking Water

Management.


Two of the water samples, both from Butler County, were more acidic than the recommended pH
for drinking water. Iron, manganese, arsenic, and lead were detected in water well samples from
Bradford and Butler Counties at levels higher than the Maximum Contaminant Levels (MCLs) set by
DEP’s Division of Drinking Water Management.32

It is important to note that while laboratory tests may not show exceeded levels for some of the
other substances, drinking water standards on which to base such determinations often do not
exist, including for methane, bromide, sodium, strontium, or Total Suspended Solids (TSS).

More than half of the project water samples contained methane. Although some groundwater can
contain low concentrations of methane under normal conditions, its presence could also indicate
natural gas migration from improperly cased or damaged gas wells. In addition, a recent analysis of
U.S. Geological Survey water monitoring data for an aquifer near
Pavillion, Wyoming found that thermogenic gas (which likely
comes from shale formations), as well as chemicals associated
with hydraulic fracturing, are present—evidence that strongly
suggests that these substances can seep
into water supplies following fracturing.33

Concentrations of some metals such as manganese and iron may
be elevated in Pennsylvania surface waters and soils either
naturally or due to past industrial activities, and levels can vary
regionally and seasonally.34 In 2012, Pennsylvania State University Drinking water wells have been
(PSU) researchers found that some drinking water wells in the contaminated with methane, chemicals,
state contained elevated concentrations of certain contaminants and other substances.
prior to any drilling in the area.35 For example, PSU researchers Photo by: Nadia Steinzor
found that 27 percent of pre-drilling water samples had
manganese above the DEP drinking water standard.36 In this
project, 7 out of the 9 water supplies sampled (78 percent) had manganese levels above the state
MCL; this is a much higher percentage than the PSU study. If there was no impact from drilling, one
would expect that fewer than three of our project samples would have had manganese above the
MCL.

Even when metals in ground water are naturally occurring or pre-date gas development, drilling
and hydraulic fracturing have the potential to mobilize substances in formations such as Marcellus
Shale, which is enriched with barium, uranium, chromium, and zinc and other metals.37 Also,
drilling can cause physical and chemical changes to groundwater aquifers that may result in
elevated metals and sediment concentrations in drinking water.38 In the PSU study, there were
three cases where wells within within 3,000 feet of the nearest Marcellus gas well experienced
changes in manganese, iron and sediment after drilling occurred. For example, each water well
had pre-drilling manganese concentrations near or below the drinking water standard (0.05 mg/L)
that increased far above the standard following drilling.39


SYMPTOM AND TESTING ASSOCIATIONS
More research would be required to state “cause and effect”
connections between the chemicals present in air and water in Many of the chemicals detected
specific locations and symptoms reported by particular residents. through testing are known to be
Nonetheless, associations can be made, as many of the chemicals linked to both oil and gas
detected through testing are known to be linked to both oil and gas operations and with the health
operations and with the health symptoms reported in the surveys.40 symptoms reported in the
surveys.
The air tests together detected 19 chemicals that may cause sinus,
skin, ear/nose/mouth, and neurological symptoms, 17 chemicals
that may affect vision/eyes, 16 that may induce behavioral effects,
11 that have been associated with liver damage, 9 with kidney
damage, and 8 associated with digestive/stomach problems. In
addition, the brain and nervous system may be affected by 5
chemicals that were detected, the cardiological system by 5,
muscles by 2, and blood cells by 2.41

More specifically, benzene, toluene, ethylbenzene, xylene,
chloromethane, trichloroethene, and acetone were detected at
project sites where residents reported associated symptoms in
health surveys, including in the categories of sinus/respiratory, skin,
vision/eyes, ear/nose/mouth, and neurological. Some of these
chemicals, as well as others (such as carbon tetrachloride and
tetrachloroethylene), were found at sites where survey participants
reported associated symptoms in the categories of digestion, kidney
and liver damage, and muscles. (For a full list of health symptoms
associated with the chemicals detected, see
http://health.earthworksaction.org.)
The many stages of gas development create
As shown in Table 9, 68 percent of the respondents at households multiple pathways for exposure to air and
where chemicals were detected reported symptoms known to be water pollution, such as emissions, spills,
leaks from compressors, impoundments, and
associated with those chemicals. Fayette and Washington Counties other facilities.
had the highest rate of association, followed by Greene, Bedford,
Photos by: Nadia Steinzor (top); Frank Finan (bottom)
and Butler. The total number of symptoms reported by individual
participants ranged from 2-111, but more than half of
participants reported having over 20 symptoms and nearly one-
quarter reported over 50. The highest number of number of
symptoms in households where we conducted air testing were
reported by a 26 year-old female in Fayette County (90 symptoms)
and a 51 year-old female in Bradford County (94 symptoms).


Table 9. Match between health symptoms reported by individuals at air testing sites and known effects of
chemicals detected

County Number of individuals Association between known effects of
surveyed at homes where chemicals detected and symptoms reported
testing was conducted

Average Range

Overall 65 68% 33 - 100%
Fayette 17 73% 33 - 100%
Washington 15 73% 33 - 100%
Bradford 10 58% 16 - 100%
Butler 12 63% 56 - 68%
Bedford 6 69% 63-100%
Elk 2 64% 53 - 74%
Clearfield 1 None None
Greene 1 70% 70%
Susquehanna 1 50% 50%

In addition, as shown in Table 10, the percent of individuals reporting particular types of symptoms
that are associated with chemicals detected in the air testing was generally consistent across
counties.

Table 10. Percent of individuals at air testing sites reporting symptoms associated with chemicals
detected at those sites, by symptom category
Symptom All Bedford Bradford Butler Fayette Washington Others
Category *
Vision/eyes 73 -- 100 63 69 67 60
Digestive/stomach 69 50 63 88 75 80 --
Skin reactions 63 50 63 88 69 53 40
Neurological 60 50 88 75 44 53 60
Behavioral/mood/e 54 67 50 63 63 47 40
nergy
Ear/nose/mouth 33 50 -- 38 44 33 20
Muscle problems -- -- -- -- -- 40 --
* This includes air samples from Clearfield, Elk, Greene, and Susquehanna Counties

As mentioned above, iron, manganese, arsenic, and lead were detected in water samples at levels
above Pennsylvania drinking water standards. These substances are known to be associated with


numerous symptoms reported by individuals living in the homes where those tests were
conducted, including in the categories of sinus/respiratory, skin reactions, digestive/stomach,
vision/eyes, ear/nose/mouth, neurological, muscle/joint, behavioral/mood/energy, and liver and
kidney damage. In addition, survey participants in the homes where
water test results showed the presence of methane reported health
symptoms known to be associated with the gas, including in the Where water test results
categories of sinus/respiratory, digestive/stomach, neurological, and showed the presence of
behavioral/mood/energy. methane, participants
reported health
Even though many participants indicated they had concerns with symptoms known to be
both water and air, the different types of testing conducted at
associated with the gas.
different households provides a way to explore whether there might
be particular symptoms more commonly associated with one type of
exposure. As indicated in Figure 4, there were notable differences in
several of the top symptoms reported at households where water versus air was tested, and
among survey participants as a whole. Participants with water tests had a higher occurrence of skin
rashes, difficulty breathing, skin irritation, diarrhea, persistent skin problems and sores that
wouldn’t heal, as well as a lower occurrence of severe headaches, throat irritation, and sleep
disturbances, than those with air tests and all respondents.

Figure 4: Differences in symptoms based on respondents with water and air tests


2.5. CONCLUSIONS
The data gathered through this project point to three central conclusions: (1) contaminants that
are associated with oil and gas development are present in air and water in areas where residents
are experiencing health symptoms consistent with such exposures; (2) there is a strong likelihood
that residents who are experiencing a range of health problems would not be if widespread gas
development were not occurring; and (3) by permitting widespread gas development without fully
understanding its impacts to public health—and using that lack of knowledge to justify regulatory
inaction—Pennsylvania and other states are risking the public’s health.

This project documented health symptoms and the presence of air contaminants at longer
distances from gas facilities than in other locations where similar projects have been conducted.42
This could be because the previous air testing was conducted in a limited geographical area very
close to facilities, while the surveys and testing in Pennsylvania took place in areas where wells and
facilities are more spread out. This could also help to explain why Pennsylvania residents who
don’t have gas facilities located on their own properties often report health problems and
indicates that air contaminants and odors can travel further than might have previously been
assumed.

Because of the short-term nature of the air canister testing (24 hours) and the single water tests
conducted at households, our results reflect conditions at particular “moments in time.” Factors
such as the stage of drilling, weather conditions, wind speeds, topography, geology, and whether
facilities are in operation or shut down may have an impact on the testing results. In addition,
some chemicals may have been present in water or air below detection
limits or prior to when the tests were conducted, meaning that other
exposures may have also occurred that caused the reported symptoms. More continuous testing
Given this, more continuous testing over longer periods of time and at over longer periods of
additional locations would likely reveal different chemicals, chemical time and at additional
concentrations, and associations with health impacts. locations would likely
reveal different
A related consideration for future research is the wide variation of results, chemicals, chemical
and therefore conclusions about the presence and levels of chemicals, that concentrations, and
occur depending on the laboratory used. This project used three associations with health
laboratories to supply canisters and analyze samples (a step taken to impacts.
compare the capacity and protocols of labs and to have “back up” should
one of the labs have proven inadequate). However, while all the labs tested
for the same core suite of chemicals, testing for other chemicals and the
reporting limits for detection varied. In addition, the labs did not all analyze the
samples for the same VOCs. For example, only Columbia Analytical analyzed for
Acetone and 2-Butanone, while only Pace Analytical analyzed for n-Hexane and
Total Hydrocarbons as gas.

More research is warranted to establish connections between reported health
problems and particular events related to gas operations, such as chemical Photo by: Frank Finan
spills, leaking waste pits, and flaring and venting. This could include, for
example, examination of case files compiled by regulatory agencies, interviews
with residents near the facilities where problems occurred, and daily odor and symptom logs kept
by residents.


EARTHWORKS Study: Gas Patch Routlette - How Shale Gas Development Risks Public Health in Pennsylvania

EARTHWORKS Study: Gas Patch Routlette - How Shale Gas Development Risks Public Health in Pennsylvania

Recommended

Recommended

More Related Content

Similar to EARTHWORKS Study: Gas Patch Routlette - How Shale Gas Development Risks Public Health in Pennsylvania

Similar to EARTHWORKS Study: Gas Patch Routlette - How Shale Gas Development Risks Public Health in Pennsylvania (20)

More from Marcellus Drilling News

More from Marcellus Drilling News (20)

Recently uploaded

Recently uploaded (13)

EARTHWORKS Study: Gas Patch Routlette - How Shale Gas Development Risks Public Health in Pennsylvania